Hi, if I want to use the moment selection criteria proposed by Hall,
A. R., Inoue, A., Jana, K. and Shin, C. (2007). Information in
generalized method of moments estimation and entropy-based moment
selection, Journal of Econometrics 138(2): 488–512.
So the criteria is computed as:
RMSC(var.xxx) = determinant(vcov, logarithm=T) +
(det(var.xxx)-p)*log(T^(1/3)/T^(1/3)
where vcov is the covariance matrix of the model parameters
conditional on the instrument set var.xxx, T the sample size and p the
number of parameters to be estimated.
Is is wrong?
Thanks,
Sebastián.
2009/1/12 Eric Zivot <ezivot@u.washington.edu>:
> You have to code these yourself
>
> -----Original Message-----
> From: Sebastian Kruk [mailto:residuo.solow@gmail.com]
> Sent: Monday, January 12, 2009 2:10 AM
> To: Eric Zivot
> Subject: Re: [S] GMM: Underidentification, weak identification, and
> instrument
>
> I have student version'S+FinMetrics but I doesn't have weak instrument tests
> like Kleibergen and Paap or any kind of test for GMM fittings.
>
> Bye,
>
> Sebastián.
>
> 2009/1/11 Eric Zivot <ezivot@u.washington.edu>:
>> I suggest you look at the recent paper by Frank Kleibergen and
>> Sophocles Mavroeidis "Weak Instrument Robust Tests in GMM and the New
>> Keynesian Phillips Curve," forthcoming in the Journal of Business and
>> Economic Statistics. A preliminary version of this paper can be
>> downloaded from Frank's webpage at Brown University Economics dept. Be
>> warned - the weak instrument procedures require the continuous
>> updating GMM estimator (which is implemented in the GMM function in
>> S+FinMetrics)
>>
>>
>> ****************************************************************
>> * Eric Zivot *
>> * Professor and Gary Waterman Distinguished Scholar *
>> * Department of Economics *
>> * Adjunct Professor of Finance *
>> * Adjunct Professor of Statistics
>> * Box 353330 email: ezivot@u.washington.edu *
>> * University of Washington phone: 206-543-6715 *
>> * Seattle, WA 98195-3330 *
>> *
>> * www: http://faculty.washington.edu/ezivot *
>> ****************************************************************
>>
>> On Sun, 11 Jan 2009, Sebastian Kruk wrote:
>>
>>> Hi,
>>>
>>> I would like to know how to test Underidentification, weak
>>> identification, and instrument relevance in GMM.
>>>
>>> Thanks.
>>>
>>> Sebastian.
>>> --------------------------------------------------------------------
>>> This message was distributed by s-news@lists.biostat.wustl.edu. To
>>> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu
>>> with the BODY of the message: unsubscribe s-news
>>>
>>
>>
>>
>
>
|